No Arabic abstract
We present a novel fitting formula for the halo concentration enhancement in chameleon $f(R)$ gravity relative to General Relativity (GR). The formula is derived by employing a large set of $N$-body simulations of the Hu-Sawicki $f(R)$ model which cover a wide range of model and cosmological parameters, resolutions and simulation box sizes. The complicated dependence of the concentration on halo mass $M$, redshift $z$, and the $f(R)$ and cosmological parameters can be combined into a simpler form that depends only on a rescaled mass $M/10^{p_2}$, with $p_2equiv1.5log_{10}left[|{bar{f}_R(z)}|/(1+z)right]+21.64$ and $bar{f}_R(z)$ the background scalar field at $z$, irrespective of the $f(R)$ model parameter. Our fitting formula can describe the concentration enhancement well for redshifts $zleq3$, nearly 7 orders of magnitude in $M/10^{p_2}$ and five decades in halo mass. This is part of a series of works which aims to provide a general framework for self-consistent and unbiased tests of gravity using present and upcoming galaxy cluster surveys. The fitting formula, which is the first quantitative model for the concentration enhancement due to chameleon type modified gravity, is an important part in this framework and will allow continuous exploration of the parameter space. It can also be used to model other statistics such as the matter power spectrum.
We present a Markov chain Monte Carlo pipeline that can be used for robust and unbiased constraints of $f(R)$ gravity using galaxy cluster number counts. This pipeline makes use of a detailed modelling of the halo mass function in $f(R)$ gravity, which is based on the spherical collapse model and calibrated by simulations, and fully accounts for the effects of the fifth force on the dynamical mass, the halo concentration and the observable-mass scaling relations. Using a set of mock cluster catalogues observed through the thermal Sunyaev-Zeldovich effect, we demonstrate that this pipeline, which constrains the present-day background scalar field $f_{R0}$, performs very well for both $Lambda$CDM and $f(R)$ fiducial cosmologies. We find that using an incomplete treatment of the scaling relation, which could deviate from the usual power-law behaviour in $f(R)$ gravity, can lead to imprecise and biased constraints. We also find that various degeneracies between the modified gravity, cosmological and scaling relation parameters can significantly affect the constraints, and show how this can be rectified by using tighter priors and better knowledge of the cosmological and scaling relation parameters. Our pipeline can be easily extended to other modified gravity models, to test gravity on large scales using galaxy cluster catalogues from ongoing and upcoming surveys.
We study and model the properties of galaxy clusters in the normal-branch Dvali-Gabadadze-Porrati (nDGP) model of gravity, which is representative of a wide class of theories which exhibit the Vainshtein screening mechanism. Using the first cosmological simulations which incorporate both full baryonic physics and nDGP, we find that, despite being efficiently screened within clusters, the fifth force can raise the temperature of the intra-cluster gas, affecting the scaling relations between the cluster mass and three observable mass proxies: the gas temperature, the Compton $Y$-parameter of the Sunyaev-Zeldovich effect and the X-ray analogue of the $Y$-parameter. Therefore, unless properly accounted for, this could lead to biased measurements of the cluster mass in tests that make use of cluster observations, such as cluster number counts, to probe gravity. Using a suite of dark-matter-only simulations, which span a wide range of box sizes and resolutions, and which feature very different strengths of the fifth force, we also calibrate general fitting formulae which can reproduce the nDGP halo concentration at percent accuracy for $0leq zleq1$, and halo mass function with $lesssim3%$ accuracy at $0leq zleq1$ (increasing to $lesssim5%$ for $1leq zleq 2$), over a halo mass range spanning four orders of magnitude. Our model for the concentration can be used for converting between halo mass overdensities and predicting statistics such as the nonlinear matter power spectrum. The results of this work will form part of a framework for unbiased constraints of gravity using the data from ongoing and upcoming cluster surveys.
We introduce the idea of {it effective} dark matter halo catalog in $f(R)$ gravity, which is built using the {it effective} density field. Using a suite of high resolution N-body simulations, we find that the dynamical properties of halos, such as the distribution of density, velocity dispersion, specific angular momentum and spin, in the effective catalog of $f(R)$ gravity closely mimic those in the $Lambda$CDM model. Thus, when using effective halos, an $f(R)$ model can be viewed as a $Lambda$CDM model. This effective catalog therefore provides a convenient way for studying the baryonic physics, the galaxy halo occupation distribution and even semi-analytical galaxy formation in $f(R)$ cosmologies.
We present an analysis of galaxy-galaxy weak gravitational lensing (GGL) in chameleon $f(R)$ gravity - a leading candidate of non-standard gravity models. For the analysis we have created mock galaxy catalogues based on dark matter haloes from two sets of numerical simulations, using a halo occupation distribution (HOD) prescription which allows a redshift dependence of galaxy number density. To make a fairer comparison between the $f(R)$ and $Lambda$CDM models, their HOD parameters are tuned so that the galaxy two-point correlation functions in real space (and therefore the projected two-point correlation functions) match. While the $f(R)$ model predicts an enhancement of the convergence power spectrum by up to $sim30%$ compared to the standard $Lambda$CDM model with the same parameters, the maximum enhancement of GGL is only half as large and less than 5% on separations above $sim1$-$2h^{-1}$Mpc, because the latter is a cross correlation of shear (or matter, which is more strongly affected by modified gravity) and galaxy (which is weakly affected given the good match between galaxy auto correlations in the two models) fields. We also study the possibility of reconstructing the matter power spectrum by combination of GGL and galaxy clustering in $f(R)$ gravity. We find that the galaxy-matter cross correlation coefficient remains at unity down to $sim2$-$3h^{-1}$Mpc at relevant redshifts even in $f(R)$ gravity, indicating joint analysis of GGL and galaxy clustering can be a powerful probe of matter density fluctuations in chameleon gravity. The scale dependence of the model differences in their predictions of GGL can potentially allow to break the degeneracy between $f(R)$ gravity and other cosmological parameters such as $Omega_m$ and $sigma_8$.
Modifications of the equations of general relativity at large distances offer one possibility to explain the observed properties of our Universe without invoking a cosmological constant. Numerous proposals for such modified gravity cosmologies exist, but often their consequences for structure formation in the non-linear sector are not yet accurately known. In this work, we employ high-resolution numerical simulations of f(R)-gravity models coupled with a semi-analytic model (SAM) for galaxy formation to obtain detailed predictions for the evolution of galaxy properties. The f(R)-gravity models imply the existence of a `fifth-force, which is however locally suppressed, preserving the successes of general relativity on solar system scales. We show that dark matter haloes in f(R)-gravity models are characterized by a modified virial scaling with respect to the LCDM scenario, reflecting a higher dark matter velocity dispersion at a given mass. This effect is taken into account in the SAM by an appropriate modification of the mass--temperature relation. We find that the statistical properties predicted for galaxies (such as the stellar mass function and the cosmic star formation rate) in f(R)-gravity show generally only very small differences relative to LCDM, smaller than the dispersion between the results of different SAM models, which can be viewed as a measure of their systematic uncertainty. We also demonstrate that galaxy bias is not able to disentangle between f(R)-gravity and the standard cosmological scenario. However, f(R)-gravity imprints modifications in the linear growth rate of cosmic structures at large scale, which can be recovered from the statistical properties of large galaxy samples.